Re: pre-populate zle for next command?

[Mikael Magnusson <mikachu@xxxxxxxxx>]

On Thu, Aug 31, 2017 at 7:38 PM, Stephen Talley
<stephentalley@xxxxxxxxxxxxxxxxx> wrote:
> Is there any way for a command to pre-populate the the zle for the next
> command?
>
> Suppose I have, for example, a zsh function "buildcmd" that produces a
> command line (based on supplied arguments, say) that the user would then be
> able to edit in the zle before hitting enter to accept the line and execute
> it.  The flow would be:
>
>     % buildcmd --my --args<enter>
>     % <output_from_buildcmd>
>
> I know I could just do:
>
>     % `buildcmd --my --args`<tab>
>
> to achieve the same thing, but it's a bit more tedious than I'd like.
>
> Ideally there'd be some hook (precmd?  accept-line?) that could check a
> variable and pre-populate the zle:
>
>   buildcmd() {
>     zle_prepopulate="some command to edit"
>   }
>
>   precmd() {
>     if [ -n "$zle_prepopulate" ]
>     then
>       zle -U "$zle_prepopulate"
>     fi
>   }
>
> ...but of course this doesn't quite work because the call to zle is not in
> the context of a widget.
>
> Is there a way?

You can use print -z to push any string you like on the zle editor
stack, which effectively does what you want. (Ie, it is popped when
the next command line is to be entered, which is immediately). There's
no need to involve precmd or any hook, you can just call print -z
directly from your function.

-- 
Mikael Magnusson